If you’re a technician or work in the field of telecommunications and networking, you may have heard the term
“network latency” when discussing the cloud or local servers. It’s also mentioned in networking contexts and can apply to almost any scenario involving data.
Simply put, latency is a measure of the delay between two points. In other words, it’s how long the pause is when data moves through a network.
But it doesn’t apply only to data in the specific sense. It can apply to the movement of anything between two points. For example, radio waves, sound waves, or even the movement of employees between two points.
However, it is primarily referred to when analyzing data movement and the time it takes for information to move from one point to another. This could be the time it takes for information to travel from a website to its destination or to enter data into a computer and wait for a response (such as opening an application, a file, or simply typing in a document).
When referring to
network latency, the measurement is taken by calculating the round-trip time of the data—that is, entering a command and waiting for the response to arrive.
Network latency is measured in milliseconds (ms); the lower the number, the better the latency—and, therefore, the better the network performance. However, it is difficult to assess latency without knowing the context of the measurement.
What constitutes low latency depends largely on the system being used. For example, a home Ethernet connection typically operates at around 10 ms, resulting in a noticeable drop in performance if it exceeds 150 ms. However, for 4G mobile connections, normal operations occur at around 45 ms to 60 ms, while 3G connections can take twice as long.
In an ideal world, every connection would have zero latency; however, there are so many interacting variables that this is unlikely to become a reality.
Even in a perfect scenario, the act of transferring a data packet from one node to another at the speed of light—known as propagation—will cause some delay. Furthermore, the larger the packet size, the longer it will take to travel through a network.
Infrastructure and hardware also play a role. Wired connections will produce varying degrees of latency depending on the type of line used—whether coaxial or fiber—and if the packet has to travel over a Wi-Fi connection, this will add even more delay to the process.
Network Latency vs. Bandwidth
Latency and bandwidth are not interchangeable terms; both are important for evaluating a network’s effectiveness.
Bandwidth refers to the network’s capacity. A line with high bandwidth can handle more traffic traveling simultaneously across a network. In the case of a business network, this means that more employees can perform network-related tasks at the same time.
However, this does not mean that data moves faster. To determine that, you must evaluate network latency, which must be low if you want responsive services.
In
MercadoIT We carry the Nexus family of
Cisco and the entire
Arista product line, which is specifically designed for data center environments where latency is a critical factor.
Reducing Network Latency
Given the complexity of some networks, reducing latency can be difficult to achieve with a single update. To truly reduce it, you may need to make adjustments—both large and small—to every aspect of a network through which a data packet will travel.
Upgrading or modifying the infrastructure itself is one way to reduce latency, which includes replacing older switches or cabling with more efficient alternatives. Operators can also evaluate their network designs to identify bottlenecks or servers that may require additional resources, or optimize them to reduce the number of nodes a data packet must traverse.
For companies operating across multiple regions, it’s worth considering the use of content delivery networks (CDNs). They provide dedicated routes that are typically located at the edge—and therefore closer to end users—which often significantly reduces the distance a data packet must travel. However, these services can be quite expensive, and the types of content they support are often limited, so they may not always be worth the investment.
On the other hand, it may be helpful to consider connecting your company’s infrastructure directly to a provider’s data center, essentially bypassing a cloud provider, for example. However, these are typically expensive alternatives to a standard contract and are not necessarily the best option.
Locally, you can slightly reduce network latency by uninstalling unnecessary programs that may be interfering with your connection.
We recommend keeping in mind that while high latency can sometimes render a network inoperable, it’s just as likely that poor performance is the result of a poorly designed application or substandard infrastructure. It is important to ensure that all applications or edge devices that rely on your network are running properly and are not consuming too many of your network’s resources.
